CN106095943A - Give song recitals and know well range detection method and device - Google Patents
Give song recitals and know well range detection method and device Download PDFInfo
- Publication number
- CN106095943A CN106095943A CN201610416943.1A CN201610416943A CN106095943A CN 106095943 A CN106095943 A CN 106095943A CN 201610416943 A CN201610416943 A CN 201610416943A CN 106095943 A CN106095943 A CN 106095943A
- Authority
- CN
- China
- Prior art keywords
- song
- fundamental frequency
- grade
- voice data
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/683—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/685—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using automatically derived transcript of audio data, e.g. lyrics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/683—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
Landscapes
- Engineering & Computer Science (AREA)
- Library & Information Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Telephonic Communication Services (AREA)
- Reverberation, Karaoke And Other Acoustics (AREA)
Abstract
The present invention relates to one give song recitals and know well range detection method and device.Described method includes: obtains and sings grade, chooses the song corresponding with described performance grade；Obtain user's voice data according to the described song recordings chosen；Extract the melody characteristics in the voice data of described recording, obtain the melody characteristics of user；The original melody characteristics of the melody characteristics of described user with the described song chosen is compared, obtains Similarity value；Judge that whether described Similarity value is more than the threshold value preset, if, then prompting detection terminates, described performance grade is known well as giving song recitals range grade, and give song recitals and know well range grade, if not described in exporting, then it represents that described user is by described performance grade, and obtain described performance grade adjacent next sing grade, continue cycling through execution.Achieve and detect that giving song recitals of user knows well range.
Description
Technical field
The present invention relates to computer application field, particularly relate to one and give song recitals and know well range detection method and dress
Put.
Background technology
Along with computer technology and the development of network technology, increasing user is engaged in various social alive by network
Dynamic, oneself life abundant, that the most also enjoys that network brings is convenient.Such as, conventional user is generally providing musical instruments
The song that oneself is familiar with is sung in place, current user can direct recording song upload to network, but usual joint performance of user
Sing the song oneself being familiar with, it is impossible to that understands song of oneself singing in antiphonal style knows well range.
Summary of the invention
Based on this, it is necessary to the problem that song is known well range cannot be understood for user, it is provided that one gives song recitals
Knowing well range detection method, can detect gives song recitals knows well range.
Additionally, there is a need to provide one to give song recitals to know well range detection device, can detect gives song recitals knows well extensively
Degree.
One gives song recitals and knows well range detection method, including:
Step A, obtains and sings grade, choose the song corresponding with described performance grade；
Step B, obtains user according to the described singing songs chosen the voice data recorded；
Step C, extracts the melody characteristics in the voice data of described recording, obtains the melody characteristics of user；
Step D, compares the original melody characteristics of the melody characteristics of described user with the described song chosen, obtains
Detected value；
Step E, it is judged that whether described detected value is more than the threshold value preset, the most then prompting detection terminates, by described performance
Grade knows well range grade as giving song recitals, and gives song recitals described in exporting and know well range grade, if not, then it represents that described use
Family by described performance grade, and obtain described performance grade adjacent next sing grade, continue cycling through execution step A to E.
One gives song recitals and knows well range detection device, including:
Choose module, be used for obtaining performance grade, choose the song corresponding with described performance grade；
Voice data acquisition module, for obtaining user according to the described singing songs chosen the voice data recorded；
Extraction module, the melody characteristics in the voice data extracting described recording, obtain the melody characteristics of user；
Comparing module, for comparing the melody characteristics of described user with the original melody characteristics of the described song chosen
Right, obtain detected value；
Judge module, for judging that whether described detected value is more than the threshold value preset；
Output module, for judging that described detected value terminates, by described performance more than the threshold value preset, prompting detection
Grade knows well range grade as giving song recitals, and gives song recitals described in exporting and know well range grade；
Enter module, for judging that described Similarity value, less than or equal to the threshold value preset, represents that described user is led to
Cross described performance grade, and obtain described performance grade adjacent next sing grade, continue to be chosen module, audio frequency number by described
Perform according to acquisition module, extraction module, comparing module, judge module, output module and entrance Module cycle.
Above-mentioned giving song recitals knows well range detection method and device, obtains and sings grade, chooses with to sing grade corresponding
Song, records the voice data that user carries out singing according to selected song, extracts the melody characteristics in the voice data recorded,
To user's melody characteristics, user's melody characteristics and original melody characteristics comparison being obtained detected value, detected value is more than the threshold preset
Value, then know well this performance grade range grade, and export, know well range as giving song recitals of this user as giving song recitals
Grade, if less than or equal to the threshold value preset, then continues to obtain next and sings grade, then the song choosing correspondence detects,
Until detecting that giving song recitals of this user knows well range grade, it is achieved that detect that giving song recitals of user knows well range.
Accompanying drawing explanation
Figure 1A is the internal structure schematic diagram of terminal in an embodiment；
Figure 1B is the internal structure schematic diagram of server in an embodiment；
Fig. 2 is to give song recitals in an embodiment to know well the flow chart of range detection method；
Fig. 3 is the melody characteristics in the voice data extracting this recording in an embodiment, obtains the melody characteristics of user
Particular flow sheet；
Fig. 4 is to be compared by the original melody characteristics of the melody characteristics of this user with this song chosen in an embodiment
Right, obtain the particular flow sheet of detected value；
Fig. 5 A is the structured flowchart giving song recitals in an embodiment and knowing well range detection device；
Fig. 5 B is the structured flowchart giving song recitals in another embodiment and knowing well range detection device；
Fig. 6 is the structured flowchart giving song recitals in another embodiment and knowing well range detection device；
Fig. 7 is the internal structure block diagram of extraction module in an embodiment；
Fig. 8 is the internal structure block diagram of comparing module in an embodiment；
Fig. 9 is the structured flowchart giving song recitals in another embodiment and knowing well range detection device.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, right
The present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, and
It is not used in the restriction present invention.
Figure 1A is the internal structure schematic diagram of terminal in an embodiment (or electronic equipment etc.).As shown in Figure 1A, this is whole
End includes that the processor connected by system bus, non-volatile memory medium, built-in storage, network interface, sound collection are filled
Put, speaker, display screen and input equipment.Wherein, the non-volatile memory medium storage of terminal has operating system, also includes one
Kind giving song recitals and to know well range detection device, this gives song recitals and knows well range detection device and give song recitals know well for realizing one
Range detection method.This processor is used for providing calculating and control ability, supports the operation of whole terminal.Interior storage in terminal
The operation that device knows well range detection device for giving song recitals in non-volatile memory medium provides environment, can in this built-in storage
Store computer-readable instruction, when this computer-readable instruction is performed by described processor, described processor can be made to perform
One gives song recitals and knows well range detection method.Network interface is for carrying out network service etc. with server.The display screen of terminal
Can be LCDs or electric ink display screen etc., input equipment can be the touch layer covered on display screen, it is possible to
To be button, trace ball or the Trackpad arranged in terminal enclosure, it is also possible to be external keyboard, Trackpad or mouse etc..Should
Terminal can be mobile phone, panel computer or personal digital assistant or Wearable etc..It will be understood by those skilled in the art that
Structure shown in Figure 1A, is only the block diagram of the part-structure relevant to the application scheme, is not intended that the application scheme
The restriction of the terminal being applied thereon, concrete terminal can include than shown in figure more or less of parts, or group
Close some parts, or there is different parts layouts.
Figure 1B is the internal structure schematic diagram of server in an embodiment (or high in the clouds etc.).As shown in Figure 1B, this service
Device includes processor, non-volatile memory medium, built-in storage and the network interface connected by system bus.Wherein, these clothes
The non-volatile memory medium storage of business device has operating system, data base and giving song recitals to know well range to detect device, data base
Middle storage has song, sings the corresponding relation etc. of grade, song and performance grade, and this gives song recitals and knows well range detection device use
Give song recitals know well range detection method in realizing being applicable to the one of server.The processor of this server is used for providing calculating
And control ability, support the operation of whole server.The built-in storage of this server is the performance in non-volatile memory medium
Song knows well the operation of range detection device provides environment, can store computer-readable instruction, this calculating in this built-in storage
When machine instructions is performed by described processor, described processor execution one can be made to give song recitals and to know well range detection side
Method.The network interface of this server is connected communication etc. with outside terminal by network for according to this.Server can be with independent
Server or multiple server composition server cluster realize.It will be understood by those skilled in the art that in Figure 1B
The structure illustrated, is only the block diagram of the part-structure relevant to the application scheme, is not intended that and is applied the application scheme
The restriction of server thereon, concrete server can include than shown in figure more or less of parts, or combination
Some parts, or there is different parts layouts.
Fig. 2 is to give song recitals in an embodiment to know well the flow chart of range detection method.Sing as in figure 2 it is shown, a kind of
Song knows well range detection method, including:
Step 202, obtains and sings grade, choose the song corresponding with this performance grade.
In the present embodiment, after performance grade refers to be divided into multiple performance grade by giving song recitals, the performance that user is to be detected
Grade.
Obtain performance grade k that user corresponding to ID has passed through, it is judged that whether performance grade k passed through is
Maximum performance grade, if, then it is assumed that user, by whole performance grades, terminates this detection, and prompting has been led to
Cross whole performance grades, if it is not, performance grade k passed through increased by 1 as performance grade to be detected, at random to examine
Song corresponding to performance grade k+1 surveyed is chosen one or number of songs.
Can be obtained by terminal and sing grade, choose the song corresponding with singing grade；Also can be obtained by terminal and sing grade,
And upload onto the server, by server according to the performance grade uploaded, choose the song corresponding with this performance grade, and return choosing
The song taken is to terminal.
In one embodiment, sing grade in this acquisition, before choosing the step of the song corresponding with this performance grade,
Also include: the degree of spreading of the song in acquisition music libraries；It is ranked up from high to low according to the degree of spreading of this song；After sorting
Song be divided into first quantity sing grade, each performance grade includes the second quantity song, and the song that degree of spreading is high
Performance grade belonging to song is low.
In the present embodiment, music libraries refers to the data base for storing song.The degree that the song that refers to degree of spreading is spread.
First quantity and the second quantity can set as required.First quantity uses n to represent, the second quantity uses t to represent.N and t is
Natural number.The song degree of spreading singing the lowest grade is the highest.The performance difficulty singing the lowest grade correspondence is the least.Singing grade can
1 to n is used to represent.Sing grade 1 grade and represent that performance grade is minimum, sing grade n level and represent that performance grade is maximum, it is possible to employing
Other modes represent.
In one embodiment, above-mentioned giving song recitals is known well range detection method and is also included: obtain the song in music libraries
Degree of spreading before, obtain the song that each song is corresponding in music libraries and sung quantity, song on-line time and song is listened to
Quantity；Sung quantity, song on-line time according to the song that each song is corresponding and song is listened to quantity and obtained by weighting
The degree of spreading that each song is corresponding.
Song is sung the total degree that quantity refers to that song is sung.Song on-line time refers to that song is positioned at appointment net
Duration on network platform.Song listens to the total degree that quantity refers to that song is listened to.
Sung quantity, song on-line time according to the song that song is corresponding and song is listened to quantity and obtained by weighting
The computing formula such as formula (1) of degree of the spreading f that each song is corresponding.
F (x, y, z)=a0x+a1y+a2z+a3 (1)
Wherein, x represents that song is sung quantity, and y represents song on-line time, and z represents that song listens to quantity, a0、a1、
a2、a3It is coefficient.Song is sung that quantity is the biggest, song is listened to quantity is the biggest, song on-line time is the shortest song
Circulation degree is the highest.
Step 204, obtains user's voice data according to this song recordings chosen.
In one embodiment, the voice data that user carries out singing opera arias and recording can be obtained according to the song chosen.Sing opera arias
Song selected by Shi Buhui broadcasting, is directly sung by user.
In one embodiment, before acquisition user is according to the voice data of the song recordings chosen, described choosing is play
The accompaniment of the song taken.Obtain user according to the song chosen the voice data recorded according to the accompaniment play.
In one embodiment, before acquisition user is according to the voice data of the song recordings chosen, according to choose
Playback of songs is for the prompt tone of vocal accompaniment.
The prompt tone of vocal accompaniment is for assisting user to find the lyrics and tune.According to the playback of songs chosen in terminal
For the prompt tone of vocal accompaniment, this prompt tone being used for vocal accompaniment is a part of content of selected song, such as prelude part.Playing
During prompt tone, terminal is shown the song information chosen.This song information can include song title, Ge Shouming, album name etc..
In one embodiment, include for the prompt tone of vocal accompaniment according to the playback of songs chosen: the song chosen according to this
Bent broadcasting is less than the prompt tone for vocal accompaniment of the first preset duration.
Obtain user to include according to the step of the voice data of this song recordings chosen: obtain user and be used for accompanying according to this
The voice data more than the second preset duration that the prompt tone sung is recorded.
First preset duration and the second preset duration can set as required.If the first preset duration can be 15 seconds, second
Preset duration can be 20 seconds.After song is selected out, terminal is play a part of content of song, such as prelude etc..Play
Prompt tone less than the first preset duration be to reduce the prompting too much to user, it is to avoid cause the inaccurate of detection.Obtain
Take family and carry out singing any one section of this song chosen according to prompt tone, and the data singing user are recorded, record
The voice data of system will be more than the second preset duration.The voice data recorded is in order to more more than the purpose of the second preset duration
Record the voice data that gives song recitals of user, it is to avoid the voice data of recording is too short, and subsequent detection is inaccurate.
Step 206, extracts the melody characteristics in the voice data of this recording, obtains the melody characteristics of user.
In the present embodiment, melody characteristics can include triad sequence.Each tlv triple in triad sequence includes ternary
Initial time, the note value of tlv triple and the persistent period of tlv triple of group.
Step 208, compares the original melody characteristics of the melody characteristics of this user with this song chosen, is examined
Measured value.
In the present embodiment, obtain the original melody characteristics of the song chosen, by the melody characteristics of user and the song chosen
Original melody characteristics compare, will count with the original triad sequence of the song chosen by the triad sequence of user
Calculation obtains detected value.This detected value can be error amount.Detected value can be used to whether evaluate and test user by singing grade.
Step 210, it is judged that whether this detected value, more than the threshold value preset, if so, performs step 212, drill if it is not, obtain this
Sing next performance grade that grade is adjacent, perform step 202.
In the present embodiment, the threshold value preset can set as required.
Step 212, prompting detection terminates, and this performance grade knows well range grade as giving song recitals, and exports this and drill
Singing song knows well range grade.
Step 206 can perform to step 212 in terminal or server.
Judge detected value whether more than the threshold value preset, the most then prompting detection terminates, using this performance grade as
Give song recitals and know well range grade, export this and give song recitals and know well range grade, if not, then it represents that this user is by this performance etc.
Level, and this performance grade is increased by 1, obtain next performance grade adjacent of this performance grade, continue executing with step
202 to step 210, chooses the song corresponding with next performance grade adjacent, obtains user's sound according to the song recordings chosen
Frequency evidence, extracts the melody characteristics in the voice data of this recording, obtains the melody characteristics of user, by the melody characteristics of this user
Compare with the original melody characteristics of the song chosen, obtain detected value, it is judged that whether detected value is more than the threshold value preset, if
It is that prompting detection terminates, and this performance grade knows well as giving song recitals range grade, and exports this and give song recitals and know well range
Grade, if it is not, this is sung grade increase by 1, so circulates, until prompting detection terminates, or all of performance preset
Grade is all passed through.
Above-mentioned giving song recitals knows well range detection method, obtains and sings grade, chooses the song corresponding with singing grade, record
User processed, according to the voice data of the singing songs chosen, extracts the melody characteristics in the voice data recorded, and obtains user's rotation
Rule feature, obtains detected value by user's melody characteristics and original melody characteristics comparison, and detected value more than the threshold value preset, then should
Sing grade and know well range grade as giving song recitals, and export, know well range grade as giving song recitals of this user, if little
In or equal to the threshold value preset, then continue to obtain next and sing grade, then the song choosing correspondence detects, until detect
Range grade is known well in giving song recitals of this user, it is achieved that detect that giving song recitals of user knows well range.
In one embodiment, choosing the song corresponding with singing grade can be one or many.If the song chosen is
The most first, then record the number of songs that user sings, respectively obtain respective melody characteristics.Respective melody is original with corresponding
Melody is compared, and obtains respective detected value.Respective detected value all with preset threshold ratio relatively, if respective detected value is equal
It is not more than the threshold value preset, then it represents that user passes through this performance grade.During it is to say, have chosen number of songs, these many first songs
Song all passes through, then it represents that user passes through this performance grade.
Fig. 3 is the melody characteristics in the voice data extracting this recording in an embodiment, obtains the melody characteristics of user
Particular flow sheet.As it is shown on figure 3, the melody characteristics of this user includes triad sequence.The voice data of this this recording of extraction
In melody characteristics, the step of the melody characteristics obtaining user includes:
Step 302, extracts the fundamental frequency data in the voice data of this recording.
In the present embodiment, before the fundamental frequency data in extracting the voice data recorded, can voice data be carried out regular
It is processed as specifying the fundamental frequency data of the form of sample rate precision.This appointment sample rate precision can set as required,
If sample rate is 16KB (kilobytes), sampling precision is 16bit (position).
Owing to people's frequency that vocal cord vibration produces when sounding can produce a large amount of overtone after sound channel filters, need from sound
Frequently extracting data directly shows the fundamental tone of vibration frequency of vocal band.Fundamental tone refers to the sound that frequency of vibration is minimum.Fundamental frequency data refer to
Fundamental tone data, it may include the shifting of fundamental frequency, fundamental frequency value, frame and frame length etc..Frame moves and frame length is selected as required.As frame moves as 10ms
(millisecond), frame length is 30ms.Frame moves the lap referring to before and after two frame.Frame length refers to the length of every frame.
Step 304, obtains the unusual fundamental frequency in these fundamental frequency data, and by the fundamental frequency value zero setting of this unusual fundamental frequency.
In the present embodiment, the fundamental frequency value of the previous fundamental frequency that certain fundamental frequency is adjacent is zero, an adjacent rear fundamental frequency
Fundamental frequency value is zero, and the fundamental frequency value of this fundamental frequency is not zero, then this fundamental frequency is unusual fundamental frequency.Fundamental frequency by this unusual fundamental frequency
Value is set to zero.
These fundamental frequency data are carried out medium filtering process by step 306.
In the present embodiment, it is judged that whether fundamental frequency segment length presets frame length less than first, the most directly carries out a length of base of window
The medium filtering of frequency range length, if it is not, then every frame does the first default medium filtering counted.
Specifically, the length that the fundamental frequency that during fundamental frequency segment length refers to fundamental frequency data, continuous adjacent fundamental frequency value is not zero links up
Degree.First presets frame length can set as required, as 30 frames, 35 frames etc..
Medium filtering refers to all fundamental frequencies fundamental frequency value of each fundamental frequency being set in this fundamental frequency field window
The intermediate value of some fundamental frequency value.
Step 308, is filled with the fundamental frequency that fundamental frequency value is zero processing.
In the present embodiment, length after fundamental frequency section is set to fundamental frequency section less than the fundamental frequency value of the second zero-base frequency range presetting frame length
Last frame fundamental frequency value.Zero-base frequency range is to be linked up by the fundamental frequency that continuous adjacent fundamental frequency value is zero to be formed.Second presets frame
Length can set as required, as 15 frames.
Step 310, carries out note to the fundamental frequency value after medium filtering processes and filling processes, obtains note value.
In the present embodiment, whole fundamental frequency values being carried out note, its computing formula is (2).
Wherein, x is fundamental frequency value.
Step 312, connects together continuous in time and that note value is identical point, obtain this recording voice data three
Tuple sequence, each tlv triple in this triad sequence includes the initial time of tlv triple, the note value of tlv triple and tlv triple
Persistent period.
In the present embodiment, merge note value, continuous in time and that note value is identical point is connected together, obtain recording
The triad sequence O of voice datai, wherein, O be tlv triple (s, m, l), s be tlv triple initial time (unit can be 50 milli
Second), m is the note value of this tlv triple, and l is the persistent period (unit can be 50 milliseconds) of this tlv triple.Wherein, tlv triple rise
The duration units of beginning unit of time and tlv triple can be selected as required, is not limited to these 50 milliseconds.
By extracting the triad sequence in fundamental frequency data, as the melody characteristics of user, it is simple to original melody characteristics
Comparison, convenience of calculation, and extract fundamental frequency data, eliminate overtone, improve the accuracy of comparison, fundamental frequency data are carried out simultaneously
The detection of unusual fundamental frequency and zero setting, medium filtering and zero-base frequency are filled, and remove noise, improve the accuracy of triad sequence,
It is easy to subsequent calculations.
Step 302 can perform in terminal or on server to step 312.
In one embodiment, the unusual fundamental frequency in these fundamental frequency data of this acquisition, and by the base of this unusual fundamental frequency
After the step of frequency value zero setting, above-mentioned giving song recitals is known well range detection method and is also included: obtain fundamental frequency value in these fundamental frequency data
The paragraph time of non-zero and；Judge this paragraph time and whether more than or equal to the 3rd preset duration, the most then perform this to this
Fundamental frequency data carry out the step of medium filtering process, if it is not, then prompting detection terminates, this performance grade is ripe as giving song recitals
Know range grade, and export this and give song recitals and know well range grade.
In the present embodiment, the 3rd preset duration can set as required, such as 10 seconds, and 15 seconds etc..The fundamental frequency value of continuous adjacent
The fundamental frequency of non-zero links up one paragraph of formation, uses this bout length of time representation.Paragraph by each fundamental frequency value non-zero
Time add up summation obtain fundamental frequency value non-zero the paragraph time and.Less than the 3rd preset duration, directly prompting detection terminates, and subtracts
Few data handling procedure below, saves and calculates resource and the time of calculating.
Fig. 4 is to be compared by the original melody characteristics of the melody characteristics of this user with this song chosen in an embodiment
Right, obtain the particular flow sheet of detected value.As shown in Figure 4, by the original rotation of the melody characteristics of this user with this song chosen
Rule feature is compared, and the step obtaining detected value includes:
Step 402, obtains the first original triad sequence and the subordinate sentence information of this song of this song chosen.
In the present embodiment, the first original triad sequence and the subordinate sentence of song of the song chosen can be obtained from music libraries
Information.First original triad sequence can be converted to by the midi file of song.Directly read midi file can be formed.As
Shown in table 1.
Table 1
| S (initial time) | L (persistent period) | M (note value) |
| 42432 | 328 | 71 |
| 42761 | 328 | 74 |
| 43090 | 328 | 76 |
| 43419 | 328 | 76 |
| 43748 | 328 | 74 |
| 44076 | 328 | 71 |
| 44405 | 657 | 71 |
| 45063 | 328 | 69 |
Initial time and the unit of persistent period in table 1 are 50 milliseconds, by carrying out initial time and persistent period
Regular processed in units obtains the integer value (employing round up process) of correspondence, as the persistent period 328/50 approximates 7.657/50
Approximate 13.Other modes can also use and round process.
The subordinate sentence information of song can include initial time and end time, the subordinate sentence quantity etc. of each subordinate sentence.
Step 404, according to the second original ternary that the subordinate sentence quantity of this subordinate sentence this song of acquisition of information and each subordinate sentence are corresponding
Group sequence.
In the present embodiment, during according to the initial time of tlv triple each in triad sequence and end time and subordinate sentence initial
Between and the end time compare, which subordinate sentence the initial time of tlv triple and end time fall in, then this tlv triple belongs to
This subordinate sentence.Such each subordinate sentence obtains a triad sequence, is the second original triad sequence.
Step 406, with each subordinate sentence as starting point, obtains subordinate sentence quantity the 3rd original triad sequence, and each the 3rd
The number of original tlv triple and the number of tlv triple in the triad sequence of the voice data of this recording in original triad sequence
Identical.
In the present embodiment, the voice data recorded because of user can be to start to sing from any subordinate sentence to obtain, therefore with each
Subordinate sentence is starting point, chooses the original tlv triple composition identical with the number of tlv triple in the triad sequence of the voice data recorded
3rd original triad sequence.Because there being subordinate sentence quantity subordinate sentence, then obtain subordinate sentence quantity the 3rd original triad sequence.
Step 408, between the triad sequence of the voice data calculating each the 3rd original triad sequence and this recording
Distance, choose minimum distance as optimal distance.
In one embodiment, step 408 includes: in calculating the 3rd original triad sequence, original tlv triple is with corresponding
The absolute value of the note value difference of tlv triple in the triad sequence of the voice data of this recording, adds both persistent period differences
Absolute value, obtains distance between the two；Calculate the tlv triple sequence of the 3rd original triad sequence and the voice data of this recording
Arrange the distance sum between tlv triple one to one, obtain the ternary of the 3rd original triad sequence and the voice data of this recording
Distance between group sequence.
Specifically, the triad sequence of the voice data such as recorded include three tlv triple (s11, m11, l11),
(s12, m12, l12) and (s13, m13, l13), the 3rd original triad sequence include three original tlv triple (s21, m21,
L21), (s22, m22, l22) and (s23, m23, l23), then (s11, m11, l11) and (s21, m21, l21) are corresponding, (s12,
M12, l12) corresponding with (s22, m22, l22), (s13, m13, l13) is corresponding with (s23, m23, l23).(s11, m11, l11) with
The distance of (s21, m21, l21) is L1=| m11-m21 |+| l11-l21 |, between (s12, m12, l12) Yu (s22, m22, l22)
Distance be L2=| m12-m22 |+| l12-l22 |, the distance between (s13, m13, l13) Yu (s23, m23, l23) is L3=|
M13-m23 |+| l13-l23 |, then between the triad sequence of the voice data of the 3rd original triad sequence and this recording away from
From for L=L1+L2+L3.
Step 410, by this optimal distance divided by the paragraph time of fundamental frequency value non-zero in these fundamental frequency data and, obtain error
Rate, using this error rate as detected value.
Step 402 can perform to step 410 in terminal or server.
Triad sequence above by original triad sequence with the voice data of recording is compared, obtain optimum away from
From, then by optimal distance divided by the paragraph time of fundamental frequency value non-zero and, available average distance, using this average distance as error
Rate, more accurately reflects error rate.
Further, this judges whether this detected value includes more than the step of the threshold value preset: whether judge this error rate
More than the threshold value preset.
The threshold value preset sets as required.
In one embodiment, above-mentioned giving song recitals is known well range detection method and is also included: obtains and shares instruction；According to this
Share to instruct and giving song recitals of ID and correspondence is known well range grade is shared to social platform.
In the present embodiment, what terminal acquisition user's trigger action produced shares instruction, shares instruction according to this and user is marked
Know and giving song recitals of correspondence is known well range grade and shared social platform.
ID is for unique character string etc. representing user identity.Social platform can include instant messaging application,
In microblogging, circle of friends etc. one or more.
Fig. 5 A is the structured flowchart giving song recitals in an embodiment and knowing well range detection device.As shown in Figure 5A, a kind of
Give song recitals and know well range detection device, including choosing module 502, voice data acquisition module 504, extraction module 506, comparison
Module 508, judge module 510, output module 512 and entrance module 514.Wherein:
Choose module 502 for obtaining performance grade, choose the song corresponding with this performance grade.
Voice data acquisition module 504 is for obtaining user's voice data according to this song recordings chosen.
The extraction module 506 melody characteristics in the voice data extracting this recording, obtains the melody characteristics of user.
Comparing module 508 is for comparing the melody characteristics of this user with the original melody characteristics of this song chosen
Right, obtain detected value.
Judge module 510 is for judging that whether this detected value is more than the threshold value preset.
Output module 512 is for judging that this detected value terminates, by this performance etc. more than the threshold value preset, prompting detection
Level knows well range grade as giving song recitals, and exports this and give song recitals and know well range grade.
Entrance module 514 is for judging that this detected value, less than or equal to the threshold value preset, represents that this user is by being somebody's turn to do
Performance grade, and obtain next performance grade adjacent of this performance grade, continue to be chosen module 502, voice data acquisition by this
Module 504, extraction module 506, comparing module 508, judge module 510, output module 512 and entrance module 514 circulation perform.
Above-mentioned giving song recitals knows well range detection device, obtains and sings grade, chooses the song corresponding with singing grade, record
User processed, according to the voice data of the singing songs chosen, extracts the melody characteristics in the voice data recorded, and obtains user's rotation
Rule feature, obtains detected value by user's melody characteristics and original melody characteristics comparison, and detected value more than the threshold value preset, then should
Sing grade and know well range grade as giving song recitals, and export, know well range grade as giving song recitals of this user, if little
In or equal to the threshold value preset, then continue to obtain next and sing grade, then the song choosing correspondence detects, until detect
Range grade is known well in giving song recitals of this user, it is achieved that detect that giving song recitals of user knows well range.
Fig. 5 B is the structured flowchart giving song recitals in another embodiment and knowing well range detection device.As shown in Figure 5 B, one
Planting gives song recitals knows well range detection device, including choosing module 502, voice data acquisition module 504, extraction module 506, ratio
To module 508, judge module 510, output module 512 and entrance module 514, also include playing module 503.
Voice data acquisition module 504 is additionally operable to obtain the audio frequency number that user carries out singing opera arias and recording according to the song chosen
According to.
In one embodiment, playing module 503 is for playing the accompaniment of this song chosen.Voice data acquisition module
504 are additionally operable to obtain user according to the song chosen the voice data recorded according to the accompaniment play.
In one embodiment, playing module 503 is additionally operable to obtaining user's audio frequency number according to the song recordings chosen
According to before, according to the playback of songs chosen for the prompt tone accompanied.
Playing module 503 is additionally operable to be less than the carrying for vocal accompaniment of the first preset duration according to this playback of songs chosen
Show sound.
Voice data acquisition module 504 is additionally operable to obtain user and presets more than second according to what this prompt tone was sung and recorded
The voice data of duration.
Fig. 6 is the structured flowchart giving song recitals in another embodiment and knowing well range detection device.As shown in Figure 6, a kind of
Give song recitals and know well range detection device, except include choosing module 502, voice data acquisition module 504, extraction module 506,
Comparing module 508, judge module 510, output module 512 and entrance module 514, also include parameter acquisition module 516, circulation degree
Computing module 518, circulation degree acquisition module 520, order module 522, grade classification module 524.Wherein:
For obtaining, the song that in music libraries, each song is corresponding is sung quantity to parameter acquisition module 516, song is reached the standard grade
Time and song listen to quantity.
Circulation degree computing module 518 for according to the song that each song is corresponding sung quantity, song on-line time and
Song is listened to quantity and is obtained, by weighting, the degree of spreading that each song is corresponding.
Circulation degree acquisition module 520 is for singing grade in this acquisition, before choosing the song corresponding with this performance grade,
The degree of spreading of the song in acquisition music libraries.
Order module 522 is for being ranked up from high to low according to the degree of spreading of this song.
Grade classification module 524 sings grade, each performance etc. for the song after sequence is divided into the first quantity
Level includes the second quantity song, and the high performance grade belonging to song of degree of spreading is low.
By extracting the triad sequence in fundamental frequency data, as the melody characteristics of user, it is simple to original melody characteristics
Comparison, convenience of calculation, and extract fundamental frequency data, eliminate overtone, improve the accuracy of comparison, fundamental frequency data are carried out simultaneously
The detection of unusual fundamental frequency and zero setting, medium filtering and zero-base frequency are filled, and remove noise, improve the accuracy of triad sequence,
It is easy to subsequent calculations.
As it is shown in fig. 7, in one embodiment, the melody characteristics of this user includes triad sequence；
This extraction module 506 include extraction unit 5061, zero setting unit 5062, filter unit 5063, fill unit 5064,
All possible combination in conversion unit 5065, combining unit 5066, time span acquiring unit 5067 and judging unit 5068.
Wherein:
The extraction unit 5061 fundamental frequency data in the voice data extracting this recording.
Zero setting unit 5062 is used for obtaining the unusual fundamental frequency in these fundamental frequency data, and by the fundamental frequency value of this unusual fundamental frequency
Zero setting.
Filter unit 5063 is for carrying out medium filtering process to these fundamental frequency data.
Fill unit 5064 for the fundamental frequency that fundamental frequency value is zero being filled with process.
Conversion unit 5065, for the fundamental frequency value after medium filtering processes and filling processes is carried out note, obtains
Note value.
Combining unit 5066, for being connected together by continuous in time and that note value is identical point, obtains the audio frequency of this recording
The triad sequence of data, each tlv triple in this triad sequence includes the note value of the initial time of tlv triple, tlv triple
Persistent period with tlv triple.
Time span acquiring unit 5067 is used for the unusual fundamental frequency in these fundamental frequency data of this acquisition, and by this unusual base
After the fundamental frequency value zero setting of frequency, obtain in these fundamental frequency data the paragraph time of fundamental frequency value non-zero and.
Judging unit 5068 is used for judging this paragraph time and whether more than or equal to the 3rd preset duration, the most then should
Filter unit 5063 is for carrying out medium filtering process to these fundamental frequency data, if it is not, then this output module 512 prompting detection knot
Bundle, knows well range grade using this performance grade as giving song recitals, and exports this and give song recitals and know well range grade.
As shown in Figure 8, this comparing module 508 includes the first acquiring unit 5081, second acquisition unit the 5082, the 3rd acquisition
Unit 5083, metrics calculation unit 5084 and similarity calculated 5085, wherein:
First acquiring unit 5081 is for obtaining the first original triad sequence of this song chosen and dividing of this song
Sentence quantity.
Second acquisition unit 5082 is for the second original triad sequence corresponding to each subordinate sentence obtaining this song.
3rd acquiring unit 5083, for each subordinate sentence as starting point, obtains subordinate sentence quantity the 3rd original tlv triple sequence
Row, and in each the 3rd original triad sequence in the triad sequence of the voice data of the number of original tlv triple and this recording
The number of tlv triple is identical.
Metrics calculation unit 5084 is for calculating the three of each the 3rd original triad sequence voice data with this recording
Distance between tuple sequence, chooses the distance of minimum as optimal distance.
Detected value computing unit 5085 for by this optimal distance divided by these fundamental frequency data during the paragraph of fundamental frequency value non-zero
Between and, obtain error rate, using this error rate as detected value.
This judge module 510 is additionally operable to judge that whether this error rate is more than the threshold value preset.
This metrics calculation unit 5084 is additionally operable to calculate original tlv triple in the 3rd original triad sequence and is somebody's turn to do with corresponding
The absolute value of the note value difference of tlv triple in the triad sequence of the voice data recorded, adds the exhausted of both persistent period differences
To value, obtain distance between the two；And
Calculate the triad sequence tlv triple one to one of the 3rd original triad sequence and the voice data of this recording
Between distance sum, the distance between the triad sequence of the voice data obtaining the 3rd original triad sequence and this recording.
Fig. 9 is the structured flowchart giving song recitals in another embodiment and knowing well range detection device.As it is shown in figure 9, it is a kind of
Give song recitals and know well range detection device, except include choosing module 502, voice data acquisition module 504, extraction module 506,
Comparing module 508, judge module 510, output module 512 and entrance module 514, also include instruction acquisition module 526 and share
Module 528.Wherein:
Instruction acquisition module 526 shares instruction for acquisition.
ID and giving song recitals of correspondence are known well range ranking score for sharing instruction according to this by sharing module 528
Enjoy to social platform.
In other embodiments, one gives song recitals and knows well range detection device, it may include chooses module 502, play mould
Block 503, voice data acquisition module 504, extraction module 506, comparing module 508, judge module 510, output module 512, enter
Enter module 514, parameter acquisition module 516, circulation degree computing module 518, circulation degree acquisition module 520, order module 522, etc.
Level divides combination the most possible in module 524, instruction acquisition module 526 and sharing module 528.
One of ordinary skill in the art will appreciate that all or part of flow process realizing in above-described embodiment method, be permissible
Instructing relevant hardware by computer program to complete, described program can be stored in a non-volatile computer and can read
In storage medium, this program is upon execution, it may include such as the flow process of the embodiment of above-mentioned each method.Wherein, described storage is situated between
Matter can be magnetic disc, CD, read-only store-memory body (Read-Only Memory, ROM) etc..
Embodiment described above only have expressed the several embodiments of the present invention, and it describes more concrete and detailed, but also
Therefore the restriction to the scope of the claims of the present invention can not be interpreted as.It should be pointed out that, for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, it is also possible to make some deformation and improvement, these broadly fall into the guarantor of the present invention
Protect scope.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (18)
1. give song recitals and know well a range detection method, including:
Step A, obtains and sings grade, choose the song corresponding with described performance grade；
Step B, obtains user's voice data according to the described song recordings chosen；
Step C, extracts the melody characteristics in the voice data of described recording, obtains the melody characteristics of user；
Step D, compares the original melody characteristics of the melody characteristics of described user with the described song chosen, is detected
Value；
Step E, it is judged that whether described detected value is more than the threshold value preset, the most then prompting detection terminates, by described performance grade
Know well range grade as giving song recitals, and give song recitals described in exporting and know well range grade, if not, then it represents that described user is led to
Cross described performance grade, and obtain described performance grade adjacent next sing grade, continue cycling through execution step A to E.
Method the most according to claim 1, it is characterised in that sing grade described acquisition, choose and described performance etc.
Before the step of the song that level is corresponding, described method also includes:
The degree of spreading of the song in acquisition music libraries；
It is ranked up from high to low according to the degree of spreading of described song；
Song after sequence being divided into the first quantity and sings grade, each performance grade includes the second quantity song, and
Performance grade belonging to the song that degree of spreading is high is low.
Method the most according to claim 2, it is characterised in that the step of the circulation degree of the song in described acquisition music libraries
Before Zhou, described method also includes:
The song that in acquisition music libraries, each song is corresponding is sung quantity, song on-line time and song listens to quantity；
Sung quantity, song on-line time according to the song that each song is corresponding and song is listened to quantity and obtained respectively by weighting
The degree of spreading that song is corresponding.
Method the most according to claim 1, it is characterised in that described acquisition user according to the described song recordings chosen
Voice data step before, described method also includes:
The prompt tone for vocal accompaniment of the first preset duration it is less than according to the described playback of songs chosen；
Described acquisition user includes according to the step of the voice data of the described song recordings chosen:
Obtain the voice data more than the second preset duration that user records according to the described prompt tone for vocal accompaniment.
Method the most according to claim 1, it is characterised in that the melody characteristics of described user includes triad sequence；
Melody characteristics in the voice data of the described recording of described extraction, the step of the melody characteristics obtaining user includes:
Extract the fundamental frequency data in the voice data of described recording；
Obtain the unusual fundamental frequency in described fundamental frequency data, and by the fundamental frequency value zero setting of described unusual fundamental frequency；
Described fundamental frequency data are carried out medium filtering process；
It is filled with the fundamental frequency that fundamental frequency value is zero processing；
Fundamental frequency value after medium filtering processes and filling processes is carried out note, obtains note value；
Continuous in time and that note value is identical point is connected together, obtains the triad sequence of the voice data of described recording,
When each tlv triple in described triad sequence includes the initial time of tlv triple, the note value of tlv triple and tlv triple lasting
Between.
Method the most according to claim 5, it is characterised in that the unusual fundamental frequency in described acquisition described fundamental frequency data
Point, and by after the step of the fundamental frequency value zero setting of described unusual fundamental frequency, described method also includes:
Obtain in described fundamental frequency data the paragraph time of fundamental frequency value non-zero and；
Judge the described paragraph time and whether more than or equal to the 3rd preset duration, the most then perform described to described fundamental frequency number
According to carrying out the step of medium filtering process, if it is not, then prompting detection terminates, described performance grade is known well wide as giving song recitals
Degree grade, and give song recitals and know well range grade described in exporting.
Method the most according to claim 5, it is characterised in that the described melody characteristics by described user is chosen with described
The original melody characteristics of song is compared, and the step obtaining Similarity value includes:
First original triad sequence of the song chosen described in acquisition and the subordinate sentence information of described song；
Subordinate sentence quantity according to song described in described subordinate sentence acquisition of information and the second original triad sequence corresponding to each subordinate sentence；
With each subordinate sentence as starting point, obtain subordinate sentence quantity the 3rd original triad sequence, and each the 3rd original tlv triple sequence
In row, the number of original tlv triple is identical with the number of tlv triple in the triad sequence of the voice data of described recording；
Distance between the triad sequence of the voice data calculating each the 3rd original triad sequence and described recording, chooses
Minimum distance is as optimal distance；
By described optimal distance divided by the paragraph time of fundamental frequency value non-zero in described fundamental frequency data and, obtain error rate, by described
Error rate is as Similarity value；
Described judge that whether described Similarity value includes more than the step of the threshold value preset:
Judge that whether described error rate is more than the threshold value preset.
Method the most according to claim 7, it is characterised in that each the 3rd original triad sequence of described calculating is with described
The step of the distance between the triad sequence of the voice data recorded includes:
Calculate the triad sequence of original tlv triple and the voice data of corresponding described recording in the 3rd original triad sequence
The absolute value of the note value difference of middle tlv triple, adds the absolute value of both persistent period differences, obtains distance between the two；
Between the triad sequence tlv triple one to one of the voice data calculating the 3rd original triad sequence and described recording
Distance sum, the distance between the triad sequence of the voice data obtaining the 3rd original triad sequence and described recording.
Method the most according to claim 1, it is characterised in that described method also includes:
Instruction is shared in acquisition；
According to described share to instruct giving song recitals of ID and correspondence is known well range grade is shared to social platform.
10. one kind give song recitals know well range detection device, it is characterised in that including:
Choose module, be used for obtaining performance grade, choose the song corresponding with described performance grade；
Voice data acquisition module, for obtaining user's voice data according to the described song recordings chosen；
Extraction module, the melody characteristics in the voice data extracting described recording, obtain the melody characteristics of user；
Comparing module, for the melody characteristics of described user is compared with the original melody characteristics of the described song chosen,
Obtain detected value；
Judge module, for judging that whether described detected value is more than the threshold value preset；
Output module, for judging that described detected value terminates, by described performance grade more than the threshold value preset, prompting detection
Know well range grade as giving song recitals, and give song recitals described in exporting and know well range grade；
Enter module, for judging that described detected value, less than or equal to the threshold value preset, represents that described user is by described
Performance grade, and obtain next performance grade adjacent of described performance grade, continue to be chosen module, voice data acquisition by described
Module, extraction module, comparing module, judge module, output module and entrance Module cycle perform.
11. devices according to claim 10, it is characterised in that described device also includes:
Circulation degree acquisition module, for singing grade described acquisition, before choosing the song corresponding with described performance grade, obtains
Take the degree of spreading of song in music libraries；
Order module, for being ranked up from high to low according to the degree of spreading of described song；
Grade classification module, sings grade for the song after sequence is divided into the first quantity, and each performance grade includes
Second quantity song, and the high performance grade belonging to song of degree of spreading is low.
12. devices according to claim 11, it is characterised in that described device also includes:
Parameter acquisition module, for obtain the song that in music libraries, each song is corresponding sung quantity, song on-line time and
Song listens to quantity；
Circulation degree computing module, is used for being sung quantity, song on-line time according to the song that each song is corresponding and song is received
Quantity is listened to obtain, by weighting, the degree of spreading that each song is corresponding.
13. devices according to claim 10, it is characterised in that described device also includes:
Playing module, for before described acquisition user's voice data according to the described song recordings chosen, according to described
The playback of songs chosen is less than the prompt tone for vocal accompaniment of the first preset duration；
Described voice data acquisition module be additionally operable to obtain user according to described for vocal accompaniment prompt tone record more than second
The voice data of preset duration.
14. devices according to claim 10, it is characterised in that the melody characteristics of described user includes triad sequence；
Described extraction module includes:
Extraction unit, the fundamental frequency data in the voice data extracting described recording；
Zero setting unit, for obtaining the unusual fundamental frequency in described fundamental frequency data, and puts the fundamental frequency value of described unusual fundamental frequency
Zero；
Filter unit, for carrying out medium filtering process to described fundamental frequency data；
Fill unit, for the fundamental frequency that fundamental frequency value is zero being filled with process；
Conversion unit, for the fundamental frequency value after medium filtering processes and filling processes is carried out note, obtains note value；
Combining unit, for being connected together by continuous in time and that note value is identical point, obtains the voice data of described recording
Triad sequence, each tlv triple in described triad sequence include the initial time of tlv triple, the note value of tlv triple and
The persistent period of tlv triple.
15. devices according to claim 14, it is characterised in that described extraction module also includes:
Time span acquiring unit, for the unusual fundamental frequency in described acquisition described fundamental frequency data, and by described unusual base
After the fundamental frequency value zero setting of frequency, obtain in described fundamental frequency data the paragraph time of fundamental frequency value non-zero and；
Judging unit, is used for judging the described paragraph time and whether more than or equal to the 3rd preset duration, the most described filtering
Unit is for carrying out medium filtering process to described fundamental frequency data, if it is not, the prompting detection of the most described output module terminates, by described
Sing grade and know well range grade as giving song recitals, and give song recitals described in exporting and know well range grade.
16. devices according to claim 14, it is characterised in that described comparing module includes:
First acquiring unit, the first original triad sequence of the song chosen described in obtain and the subordinate sentence letter of described song
Breath；
Second acquisition unit, for according to the subordinate sentence quantity of song described in described subordinate sentence acquisition of information and each subordinate sentence corresponding second
Original triad sequence；
3rd acquiring unit, for each subordinate sentence as starting point, obtains subordinate sentence quantity the 3rd original triad sequence, and each
The number of original tlv triple and tlv triple in the triad sequence of the voice data of described recording in 3rd original triad sequence
Number identical；
Metrics calculation unit, for calculating the tlv triple sequence of each the 3rd original triad sequence and the voice data of described recording
Distance between row, chooses the distance of minimum as optimal distance；
Similarity value computing unit, was used for described optimal distance divided by the paragraph time of fundamental frequency value non-zero in described fundamental frequency data
With, obtain error rate, using described error rate as Similarity value；
Described judge module is additionally operable to judge that whether described error rate is more than the threshold value preset.
17. devices according to claim 16, it is characterised in that it is original that described metrics calculation unit is additionally operable to calculate the 3rd
Original tlv triple and the note value of tlv triple in the triad sequence of the voice data of corresponding described recording in triad sequence
The absolute value of difference, adds the absolute value of both persistent period differences, obtains distance between the two；And
Between the triad sequence tlv triple one to one of the voice data calculating the 3rd original triad sequence and described recording
Distance sum, the distance between the triad sequence of the voice data obtaining the 3rd original triad sequence and described recording.
18. devices according to claim 10, it is characterised in that described device also includes:
Instruction acquisition module, shares instruction for acquisition；
Sharing module, for according to described in share instruct by ID and correspondence give song recitals know well range grade share to
Social platform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610416943.1A CN106095943B (en) | 2016-06-14 | 2016-06-14 | It gives song recitals and knows well range detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610416943.1A CN106095943B (en) | 2016-06-14 | 2016-06-14 | It gives song recitals and knows well range detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106095943A true CN106095943A (en) | 2016-11-09 |
| CN106095943B CN106095943B (en) | 2018-09-21 |
Family
ID=57846747
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610416943.1A Active CN106095943B (en) | 2016-06-14 | 2016-06-14 | It gives song recitals and knows well range detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106095943B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109087669A (en) * | 2018-10-23 | 2018-12-25 | 腾讯科技（深圳）有限公司 | Audio similarity detection method, device, storage medium and computer equipment |
| CN109754820A (en) * | 2018-12-07 | 2019-05-14 | 百度在线网络技术（北京）有限公司 | Target audio acquisition methods and device, storage medium and terminal |
| WO2020000977A1 (en) * | 2018-06-29 | 2020-01-02 | 上海掌门科技有限公司 | Method and device for generating and comparing music data and computer storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070106406A1 (en) * | 2005-10-28 | 2007-05-10 | Victor Company Of Japan, Ltd. | Music-piece classifying apparatus and method, and related computer program |
| CN102497468A (en) * | 2011-11-18 | 2012-06-13 | 上海华勤通讯技术有限公司 | Cellphone and singing self-evaluating method |
| CN104392731A (en) * | 2014-11-30 | 2015-03-04 | 陆俊 | Singing practicing method and system |
| CN105118500A (en) * | 2015-06-05 | 2015-12-02 | 福建凯米网络科技有限公司 | Singing evaluation method, system and terminal |
-
2016
- 2016-06-14 CN CN201610416943.1A patent/CN106095943B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070106406A1 (en) * | 2005-10-28 | 2007-05-10 | Victor Company Of Japan, Ltd. | Music-piece classifying apparatus and method, and related computer program |
| CN102497468A (en) * | 2011-11-18 | 2012-06-13 | 上海华勤通讯技术有限公司 | Cellphone and singing self-evaluating method |
| CN104392731A (en) * | 2014-11-30 | 2015-03-04 | 陆俊 | Singing practicing method and system |
| CN105118500A (en) * | 2015-06-05 | 2015-12-02 | 福建凯米网络科技有限公司 | Singing evaluation method, system and terminal |
Non-Patent Citations (1)
| Title |
|---|
| 刘清堂等: "基于特征比对的音乐学唱系统设计与实现", 《计算机工程与设计》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020000977A1 (en) * | 2018-06-29 | 2020-01-02 | 上海掌门科技有限公司 | Method and device for generating and comparing music data and computer storage medium |
| CN109087669A (en) * | 2018-10-23 | 2018-12-25 | 腾讯科技（深圳）有限公司 | Audio similarity detection method, device, storage medium and computer equipment |
| CN109754820A (en) * | 2018-12-07 | 2019-05-14 | 百度在线网络技术（北京）有限公司 | Target audio acquisition methods and device, storage medium and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106095943B (en) | 2018-09-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104620313B (en) | Audio signal analysis | |
| JP5565374B2 (en) | Device for changing the segmentation of audio works | |
| CN103440862B (en) | A kind of method of voice and music synthesis, device and equipment | |
| KR101612768B1 (en) | A System For Estimating A Perceptual Tempo And A Method Thereof | |
| JP4878437B2 (en) | System and method for generating audio thumbnails | |
| US8069036B2 (en) | Method and apparatus for processing audio for playback | |
| US20150340021A1 (en) | Method for extracting representative segments from music | |
| CN104143324B (en) | A kind of musical tone recognition method | |
| CN105161116B (en) | The determination method and device of multimedia file climax segment | |
| CN109299318A (en) | Method, apparatus, storage medium and the terminal device that music is recommended | |
| CN106991163A (en) | A kind of song recommendations method based on singer's sound speciality | |
| US20070157797A1 (en) | Taste profile production apparatus, taste profile production method and profile production program | |
| CN106157979B (en) | A kind of method and apparatus obtaining voice pitch data | |
| CN101740025A (en) | Singing score evaluation method and karaoke apparatus using the same | |
| CN110265064A (en) | Audio sonic boom detection method, device and storage medium | |
| US20080262836A1 (en) | Pitch estimation apparatus, pitch estimation method, and program | |
| WO2022105693A1 (en) | Sample generation method and apparatus | |
| CN106997769B (en) | Trill recognition method and device | |
| CN106095943A (en) | Give song recitals and know well range detection method and device | |
| JP4775380B2 (en) | Apparatus and method for grouping time segments of music | |
| CN106887233A (en) | Audio data processing method and system | |
| CN109920446A (en) | A kind of audio data processing method, device and computer storage medium | |
| JP4622199B2 (en) | Music search apparatus and music search method | |
| CN106782601A (en) | A kind of multimedia data processing method and its device | |
| CN106409311A (en) | Refrain extracting apparatus and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |